This R21 proposal aims to develop a risk score for cognitive complications of Type 1 Diabetes (T1D) by using longitudinal imaging markers of the retinal vasculature and state-of-the art neurocognitive assessment. Cognitive impairment (CI) is a common and severely disabling complication of T1D. With increasing incidence and survival rates in T1D, there will soon be an increase in the number of patients who are at risk of developing CI. Therefore, an urgent public health priority in T1D is to develop a biomarker that can detect brain changes before clinically overt signs of CI develop, to identify the patients at greater risk of developing CI. Although brain imaging is the gold standard to quantify neurovascular imaging abnormalities underlying CI, it is not an ideal tool for population screening because of high costs and variable access. Similarly, extensive neurocognitive batteries are not routinely administered to T1D patients and require complex adjudication procedures to classify CI with reliability. The retinal circulation offers a unique perspective on the cerebral microcirculation in vivo and retinal vascular imaging (RVI) measures are promising biomarkers of CI and neurovascular abnormalities. RVI measures have been pointed out as potential tools to increase accuracy of diagnosis of dementia in the general population of older adults. However, these associations have not been well characterized in patients with T1D. Most evidence is from cross-sectional studies of mostly young patients, and consists of associations of variable strength between retinopathy, cognitive tests and neurovascular abnormalities. This proposal marks a critical first step in the overall path toward the development and validation of RVI measures as biomarkers of CI. We propose to develop a risk score of CI using existing longitudinal RVI measures (arteriolar and venular diameter, tortuosity, nicking), neuroimaging and state of the art cognitive assessment. The proposed study leverages two existing ongoing studies of patients with T1D: the Epidemiology of Diabetes Complication Study (R37 DK034818-28, PI: Orchard), which has been following these patients since 1986-88 with repeated RVI measures, extensive characterization of health- and diabetes-related measures and a vast biorepository of blood sample; the Neurovascular Imaging Study (R01 DK089028-03, PI: Rosano) which has obtained cutting-edge neuroimaging and cognitive assessment in 2011-12. A total of 209 participants have received a 1st cognitive assessment and has retrospective repeated RVI since 1986 to time of cognitive assessment. We propose to add a 2nd cognitive assessment in a subsample of these participants who are scheduled to return for their regular visit in 2016-17. We propose to relate repeated RVI measures with neurovascular imaging abnormalities, CI and cognitive decline. This R21 application responds PA-12-15 because it proposes secondary analyses of existing data, combined with additional new data collection; the results of this project will inform and support subsequent applications to further our understanding of the pathogenesis of cognitive complications in middle-aged patients with T1D and design future intervention strategies.